Longitudinal associations between exposure to anesthesia and neurocognitive functioning in pediatric medulloblastoma

Abstract

Aim: To examine whether anesthesia exposure is associated with neurocognitive decline in pediatric medulloblastoma.

Methods: Patients were treated at St. Jude Children's Research Hospital and completed ≥2 protocol-directed neurocognitive assessments (n = 107) as part of a multisite clinical trial for pediatric medulloblastoma (NCT00085202). Patients received risk-adapted craniospinal photon irradiation, followed by four cycles of high-dose chemotherapy and stem cell rescue. Neurocognitive testing was completed at study baseline (after surgery and <2 weeks of starting radiation therapy) and annually for 5 years. Data on anesthesia exposure during treatment was abstracted from medical records.

Results: Patients were 10.2 years at diagnosis on average (SD = 4.5; 37% female, 73% average-risk). Mean cumulative anesthesia duration was 20.4 h (SD = 15.2; range 0.7-55.6 h). In the overall group, longer anesthesia duration was associated with greater declines in IQ (Estimate = -0.08, P < 0.001), attention (Estimate = -0.10, P < .001) and processing speed (Estimate = -0.13, P < 0.001). Similar results were shown in subgroups of patients who were <7 years at diagnosis (IQ = -0.14, P = 0.027; Attention = -0.25: P = 0.011), ≥7 years at diagnosis (Attention = -0.07, P = 0.039; Processing Speed = -0.08, P = 0.022), treated for high-risk disease (IQ = -0.09, P = 0.024; Attention = -0.11, P = 0.034; Processing Speed = -0.13, P = 0.001), or treated for average-risk disease (IQ = -0.05, P = .022; Attention = -0.08, P = 0.011; Processing Speed = -0.10, P < 0.001).

Conclusion: Greater anesthesia exposure is a risk factor for clinically significant neurocognitive decline, in addition to factors of age at diagnosis and treatment risk arm. This result is notable as there are evidence-based strategies that can limit the need for anesthesia. Limiting anesthesia exposure, as feasible, may mitigate neurocognitive late effects, and thus, improve quality of life for survivors.

Keywords: Anesthesia; Brain tumour; Children; Longitudinal; Medulloblastoma; Neurocognitive.

Figure 1.

Association between time and age, treatment risk, and anesthesia duration on neurocognitive decline (overall group) Notes: Estimated performance on the WJ III Attention index is shown as age standardized norms (M=100, SD=15) in the overall group (n=107). Linear mixed models showed significant interactions between time and age at diagnosis (P<.001), time and treatment risk group (P=.029), and time and anesthesia duration (P<.001). Colored lines represent estimated scores for: 3, 6, 9, 12, 15, or 18 years old at diagnosis (panel a); high risk or average risk treatment (panel b); and 0, 10, 20, 30, 40, or 50 hours of cumulative anesthesia exposure (panel c).

Figure 2.

Association between time and anesthesia duration on neurocognitive decline (age groups) Notes: Estimated performance on the WJ III Attention index is shown as age standardized norms (M=100, SD=15) within the <7 years old at diagnosis group (n=27, panel a) and ≥ 7 years old at diagnosis group (n=80, panel b). Linear mixed models showed significant interactions between time and anesthesia exposure (young age: P=.011; older age: P=.039). Colored lines represent estimated scores for: 0, 10, 20, 30, 40, and 50 hours of cumulative anesthesia exposure.

Figure 3.

Association between time and anesthesia duration on neurocognitive decline (treatment risk groups) Notes: Estimated performance on the WJ III Attention index is shown as age standardized norms (M=100, SD=15) within the high risk treatment group (n=29, panel a) and average risk treatment group (n=78, panel b). Linear mixed models showed significant interactions between time and anesthesia exposure (high risk: P=.034; average risk: P=.011). Colored lines represent estimated scores for: 0, 10, 20, 30, 40, and 50 hours of cumulative anesthesia exposure.